Structural Estimation of Controlled Diffusion Processes: Methodology and Applications

The Structural Estimation of a Controlled Diffusion Process (SECDP) consists of a methodology to estimate the parameters of a controlled diffusion process. The term controlled refers to a theory about the incentives of the agent that is influencing the process in a desired manner. This theory provides not only a understanding of the forces that drive the process., the predictions from the method proposed her are often more robust than predictions from a econometric model that simply tries to find correlation patterns in the data.

We apply this technique to study customer care in the four major phone carriers in U.S. telecommunications industry (AT&T, Verizon, Sprint, and T-Mobile). Our findings shed several insights into digital customer care in this industry. First, we find a clear separation among the firms in terms of digital customer care effectiveness. Second, we find that good customer care is not merely a matter of responding to customer tweets: T-Mobile and Sprint have high response rates, but are low on effectiveness. Third, the quality of digital customer care that customers expect varies across firms: Customers of higher priced firms (e.g., Verizon and AT\&T) expect better customer care. Fourth, seemingly unrelated events (such as signing an exclusive contract with a celebrity) can influence digital customer care. These events can be firm-initiated or exogenous. Our study has important implications for managers as it can help firms determine the optimal strategy to influence customer sentiment. The study also helps firms anticipate the impact of external events on digital customer care and adjust the response strategy to accommodate these events.